Saturday, March 22, 2014

melonJS should be *All About Speed*

I've been working on melonJS since I first used it in July 2012. It's now over 1.5 years later, and we've learned a lot as a team. One particular area of melonJS that I have focused a lot of time is performance. And in this article, I will try to describe current pain points and what we are doing to make them better.

TL;DR: A super fast prototypal inheritance microlib for the modern web. https://github.com/parasyte/jay-extend

Simple Inheritance

melonJS has always used John Resig's Simple Inheritance pattern, like many of the HTML5 game engines available. While looking for things to optimize, I stumbled upon some funny code with regular expressions in the Object.extend helper function. If you've never looked at the source, you can head over to John's original article where he introduced it. The RegExp is a hack to determine if the browser can coalesce functions into a string; in other words, it is a test to determine if the browser is capable of decompiling functions. But why on earth would it need to decompile functions? Well, for syntactic sugar, of course! To save the programmer from typing prototype so often.

To see how that regular expression got there, we first have to do some code archeology. One of the best ways to illustrate is to reinvent it. All the code I will be showing here was run using jsfiddle with my browser's web console open.

Let's start with prototypal inheritance; the concept of creating a function in a prototype form (a class), then instantiating objects from that class. Here is a very simple example of a class which shows a constructor and an accessor method:

function Foobar() {
    this.num = 123;
}

Foobar.prototype.setNum = function (num) {
    return this.num = (typeof(num) === "number") ? num : this.num;
};

In this example, the function named Foobar is the class, and its body is the constructor. Then the class prototype is modified to include an accessor method. Let's instantiate an object with this class and try it out:

console.clear();

var f = new Foobar();
console.log("Expect 123:", f.num);
console.log("Expect 9000:", f.setNum(9000));
console.log("Expect 9000:", f.setNum("abc"));

If you run the code, you'll see that it does work as intended! This is the classic example of prototypal inheritance in JavaScript, and you will see it used in tons of libraries today. But as you start to add more and more methods to your class, you will be typing prototype quite a lot. A little syntactic sugar can hide prototype by iterating over a descriptor which defines the methods you want added to the prototype.

Object.prototype.extend = function (Class, descriptor) {
    for (var method in descriptor) {
        Class.prototype[method] = descriptor[method];
    }
}

function Foobar() {
    this.num = 123;
    this.str = "xyz";
}

Object.extend(Foobar, {
    "setNum" : function (num) {
        return this.num = (typeof(num) === "number") ? num : this.num;
    },
    "setStr" : function (str) {
        return this.str = (typeof(str) === "string") ? str : this.str;
    }
});

I've added a second accessor method to our Foobar class, and also created a new function called Object.extend (same name as used in Simple Inheritance, by the way) which provides the syntactic sugar we were looking for. Hooray! No more prototype in our class definitions. And here's another quick test to prove that it works:

console.clear();

var f = new Foobar();
console.log("Expect 123:", f.num);
console.log("Expect 9000:", f.setNum(9000));
console.log("Expect 9000:", f.setNum("abc"));
console.log("Expect xyz:", f.str);
console.log("Expect abc:", f.setStr("abc"));
console.log("Expect abc:", f.setStr(9000));

Now what about inheritance? In JavaScript, you can do single inheritance quite simply by replacing the class prototype with a copy of the prototype from another class. There's a function built into the language for doing exactly that: Object.create.

(Side note: Object.create is a relatively recent addition to the JavaScript language, and Simple Inheritance pre-dates its inclusion. So Simple Inheritance uses a trick involving the instantiation of the object itself with a little hack to prevent the constructor from running during the prototype creation phase. See John's article for more details on that little catch.)

In order to create the inheritance chain, we need to modify Object.extend to return an empty class whose prototype is given a copy of the base class (thanks to Object.create). Because the empty class has its own constructor, we need a way to run the user's constructor. Again, following the Simple Inheritance API, we'll adopt init as the name of the user's constructor, and simply call that if it exists.

Object.prototype.extend = function (descriptor) {
    function Class() {
        if (this.init) {
            this.init.apply(this, arguments);
        }
        return this;
    }

    Class.prototype = Object.create(this.prototype);

    for (var method in descriptor) {
        Class.prototype[method] = descriptor[method];
    }

    return Class;
}

We'll use this updated Object.extend to recreate our base class Foobar, then create a Duck class that inherits from Foobar.

var Foobar = Object.extend({
    "init" : function () {
        this.num = 123;
        this.str = "xyz";
    },
    "setNum" : function (num) {
        return this.num = (typeof(num) === "number") ? num : this.num;
    },
    "setStr" : function (str) {
        return this.str = (typeof(str) === "string") ? str : this.str;
    }
});

var Duck = Foobar.extend({
    "speak" : function () {
        return "quack";
    }
});

And then we'll write a simple test to verify it all works as I claim:

console.clear();

var d = new Duck();
console.log("Expect 123:", d.num);
console.log("Expect 9000:", d.setNum(9000));
console.log("Expect 9000:", d.setNum("abc"));
console.log("Expect xyz:", d.str);
console.log("Expect abc:", d.setStr("abc"));
console.log("Expect abc:", d.setStr(9000));
console.log("Expect quack:", d.speak());

There you go! The reinvention of Simple Inheritance is complete. Well, almost! What happens if you want to override a method, instead of just adding new ones? You can do that with the above code, sure, but you probably want to run the overridden method, as well. This is done by going back to the prototype chain, again! Let's modify Duck to append "quack" to every string set:

var Duck = Foobar.extend({
    "speak" : function () {
        return "quack";
    },
    "setStr" : function (str) {
        return Foobar.prototype.setStr.call(this, str + " quack");
    }
});

This prototype mess is how we call a method on the super class in JavaScript! Yes, it's normal. Yes, it's ugly. Yes, that's why Simple Inheritance adds more sugar in the form of this._super! The problem to recognize is that binding the instantiated object reference (this) to a method on the prototype chain is hard to do in a generic way. John's solution is creating a proxy method called _super which redirects method invocations to the same method name on the super constructor. It manages this via a series of closures to retain scope in the prototype chain.

In order to create each proxy method, Simple Inheritance adds checks to the descriptor iteration to find methods (functions). For each method found, a RegExp test (!) is run to determine whether a proxy should be added. This, at long last, answers our original question, "why does Simple Inheritance need to decompile functions?" Frankly, it doesn't need to decompile anything! But as an optimization, it won't add the proxy to methods that don't need it. And it determines which methods need the proxy by decompiling the function; looking for a string that matches _super. So if _super is ever used in the method, it gets a proxy that adds the _super property!

"But isn't decompiling functions slow?" ... Yeah, probably!

"And isn't running a regular expression over a long function-as-a-string also slow?" ... I would say so!

*sigh*

And so began my adventure to make Simple Inheritance fast! Actually, I did so much of my own work here to reinvent Simple Inheritance that I'm just going to release it on its own, with a nod to John Resig for the inspiration.

I made _super fast by adding the proxy within the Class constructor, rather than during prototype creation time. This means that none of the methods get proxied automatically; instead, _super is the proxy. After going through multiple failed tests, I found the best interface for _super is actually closer to Python than Simple Inheritance. Notice that the Python super() function takes two arguments; a reference to the super class, and a reference to the instantiated object.

All that means is _super in Jay Inheritance looks like this:

var B = A.extend({
    "foo" : function (arg1, arg2) {
        this._super(A, "foo");
        this._super(B, "bar", [ arg1, arg2 ]);
    },
    "bar" : function (arg1, arg2) {
        // ...
    }
}

Quite a bit different from Simple Inheritance! And there is also a big difference between my final implementation and Python's super() function, which returns a proxy to the super class. If I were to do that, it would require another iteration, which I'm not willing to do, for the sake of raw performance. So instead, my _super interface accepts a reference to the super class, and a method name, followed by an array of arguments that will be proxied to the super class method.

Requiring a reference to the super class also allows calling sibling methods on the class, without going through the normal prototype chain. This is useful for base classes where you don't want to call overridden methods. Simple Inheritance does not have this functionality in syntactic sugar form; you'll have to use the ugly prototype mess for that.

As an added bonus, I also include mixin support, which allows adopting methods from a class without inheriting its prototype.

Finally, I've wrapped all of this knowledge into a single project called Jay Inheritance, named after yours truly. Here is the official gist: https://gist.github.com/parasyte/9712366 and I also have the same code on jsfiddle, where the tests can be run directly: http://jsfiddle.net/Qb7e8/6/ (Update, July 2015: the project has been release as jay-extend: https://github.com/parasyte/jay-extend and is available through dependency managers npm and bower.)


Jay Inheritance is as screaming fast as classical inheritance patterns can get in JavaScript. From here, we need to refactor melonJS classes to use it, and also to make the classes much smaller! Objects with too many properties are hard for JavaScript engines to optimize. For example, the V8 engine in Chrome switches to slow-mode when an object has about 30 properties. Source: http://youtu.be/XAqIpGU8ZZk?t=25m07s Property access also uses type-specialized optimizations, so we should not be changing the data type stored in any properties, nor adding or removing properties outside of the constructor.

In the next article, I'll talk about some modifications that were recently made to Simple Inheritance in melonJS, and why this was a bad idea from a performance perspective. I'll also go over some ideas that we can use to improve the situation in the future with the inheritance pattern.

Thursday, December 12, 2013

Ludum Dare #28; "I'm in!"

I've posted about LD48 on other forums, but I want to write a quick note about it here on the blog, as well. Maybe it will help catch a bit more attention, too. Every little bit helps! ;)

Ludum Dare #28 is starting this weekend! I will be participating, and streaming my progress live on Twitch for the full 48-hour duration. Everyone is invited to join and get involved! I'll be accepting suggestions and ideas live. (I cannot accept any resources like graphics, code, music, or sound effects, as that is against the compo rules.)

I will be making an HTML5 game using melonJS, which I will be semi-frequently publishing to github. This means you will be able to load the game and give it a try as I'm making it. I'll get instant feedback and play testing, and you might get to see your suggestions added in minutes! That's pretty awesome.

You can catch the live stream on Twitch starting Friday evening at 6:00pm Pacific (GTM -0800). The link is:

http://www.twitch.tv/kodewerx

Also, be sure to follow me on Twitter, so you can keep up-to-date on any project news and streaming announcements: @kodewerx

I have my streaming setup tested and ready to go! No telling what I will make yet. It will probably be something totally random, based on whatever theme is chosen. In any case, thanks to everyone who has said they will drop in! I'm really excited about this. It will be my first Ludum Dare, and first live stream event. I'm thinking it will be full of unintentional hilarity!

See you Friday!

Monday, December 2, 2013

A Walk in the Park Turbo: #1GAM November and #CharityGameJam II Turbo

A Walk in the Park Turbo

(Clicky for Make Big)

It's been a looooooong time since my last #1GAM entry. This time around, I started making a classic text adventure; a project I've been wanting to do for at least month. I was heavily inspired by a documentary called Get Lamp.

The game is called "A Walk in the Park Turbo", as it lovingly shares the moniker for #CharityGameJam II Turbo. The Charity Button (visible in the animation) points the player directly to the donation page for Child's Play Charity. Don't forget to make a contribution!


About the Game

"A Walk in the Park Turbo" follows in the footsteps of the earliest video games, which can be classified under the genre "interactive fiction". If you've ever read a "choose your own adventure" book, you know exactly how these games play; the player is thrown into a scenario, and left to her own devices to discover the correct actions to claim victory.

The classic text adventures are remarkably minimalistic, but can be surprisingly deep. AWITPT was made in just two days. And though there is not a whole lot to actually do, the biggest parts of the game were made near the end of that two-day sprint. Most of the development time went into building the core foundation. Which is a shockingly simple JSON data structure that the game dynamically modifies as you make your way through the puzzles.

How to Play

Text adventures are notorious for being difficult to get a grasp on. Especially for beginners. Because the user interface is all about typing commands, you'll find yourself stuck often until you can figure out which commands the programmer expects you to type ... and sometimes even in the proper order!

The general recommendation to get started is to try typing different words, and see what happens. In AWITPT, the commands are almost always verbs (but they can be shortened in some cases). There are movement commands, like "go" or "walk" which also expect a direction. Or you might be able to shorten it to just the direction. Another common command is "look" or "search" to look at your surroundings, and at items that may be available in your current location. You can also take items, and use items, check your bag (inventory), and inspect or use items in your bag. Items in your bag can also be used on some of the stuff you see in the area.

Puzzles

Text adventures are all about puzzles; How to find the right items, and use them correctly to unlock new paths. AWITPT is no different! To date, there are only 6 available areas, with most of them serving a purpose of gaining one or more items. Getting into two of those area requires using one of the items you can collect. ;)

Music and Sound Effects

I composed the music many years ago, originally as a soundtrack for a short film idea I had, called "Milksharks". The concept was a bit absurd and comedic; tiny sharks that swim around in your cereal bowl. The film never got started, but I was inspired enough to create a short music track for it. It's a calm tune that starts with an upbeat cello, and evolves over to other instruments.

The soundtrack is available on Soundcloud.

All of the sound effects (the punches and kicks on the logo) were sourced from CC0 (public domain) sounds made available on freesound.org.

The Future of A Walk in the Park Turbo

Like I mentioned at the beginning of the article, this genre is one I've really wanted to try my hand at. So I'm glad I've finally got the basis for it worked out. I'm sure I'll come back to this one to add more areas and stuff to do. I don't actually have any story elements planned for it. I want to just let it evolve naturally. In a very silly, almost dream-like game world. The best part about making a text adventure is that anything I can imagine can be put into the game. As realistic or as fantastic as I like. This game is truly a medium for artistic expression.

Apart from adding the more areas to explore, I also want to add some pretty visual effects. The game is written with a 2D canvas, so why not use it! Audio is another thing I want to highlight. Both of these enhancements will feel very much out of place in an old school text adventure. But that's really what I'm going for, here! Bringing a classic genre up to modern standards. The addition of the "Turbo" subtitle kind of sets that stage, both in name and in its very introduction on the title screen.

Make Games ...

Play online : http://blipjoy.github.io/awitp/public/index.html
Source code : https://github.com/blipjoy/awitp
#CharityGameJam : http://www.charitygamejam.com/
#1GAM : http://www.onegameamonth.com/kodewerx
Twitter : @kodewerx

With that, I will leave you with my traditional closing statement;

Make Games, Support Charity!

Wednesday, May 1, 2013

1GAM April: Postmortem

Sprung Fever

(Clicky for Make Big)

The game is officially complete! I just added the last remaining bits (item sprites, courtesy Matwek!) and I also released the game earlier today on #1GAM.

And now it's postmortem time! YAY!

Development Process

melonJS 0.9.7 is a real pleasure to work with! This game ships with a build that is about as close to final 0.9.7 release as possible (there may be a patch or two that gets committed between now and release, but it's effectively the same). The new features that I made extensive use of this time were:

  • TexturePacker support
  • JSON TMX map support
  • me.Renderable
  • me.sys.isMobile
  • me.Viewport.shake
  • TMX settings JSON parsing
During development, plenty of bugs were discovered and fixed, which is nice for the melonJS community! I'm super pleased with how melonJS 0.9.7 has turned out!


What Went Right

Wow! What can I say? Both artists did great on the assets they were able to deliver, and it all works together as a cohesive whole! Great job and a huge thanks to both Matwek and Burcag!

On the coding side, everything went rather smooth... There's lots of fancy stuff happening behind the scenes and on screen, so keep a look out for it! The fullscreen overlays worked really well, and they were incredibly easy to build. Girl AI is basic, but it gets the job done. The particle emitters are flawless. Mobile control works well (if a bit lacking in tactile response). And the sound effects make me really happy.

My favorite of all sound effects is the heartbeat. But it's so bassy, you can only hear it with good headphones or a subwoofer. That's actually a plus, as well, since playing the game with a good sound system is PHENOMENAL. And even though the music is completely auto-generated (with some tweaks) it's still catchy enough to get stuck in my head.

The gameplay also turned out to be incredibly fun. A coworker asked if she could play it last week, so she became the first play tester in the office. Her reaction was priceless. That's when I really knew it was a good game. It's definitely the most fun of all my projects so far.


What Went Wrong

Even though I have been working with two artists this time, I still feel like getting art assets has been troublesome. The girl in the blue dress was never animated for up/down movement, and the environments feel a bit repetitive/on a grid. (If you've played Neverwell Moor, you'll see what I'm talking about; that game hides the grid quite well, and it's all thanks to the art assets!)

Music was VERY difficult. I spent a lot of time just screwing around until I had something that I liked. If I had more patience with it, I would teach myself to make chiptune-sounding music in GarageBand. Or else I get music that sounds pretty nice, but doesn't fit the old-school theme. :(

This project sounds pretty simple on paper, but in reality it was a lot of work! There are just so many variables, like player control, collectable items, game stats (HUD), enemy AI, etc. But the biggest time consumer for me was probably mapping. Definitely not my favorite thing in the world... And it's frustrating to be working on a map and realize half way into it that you're missing a rather vital tile. You might notice some funny looking shadows on the left and right side of the grocery map! Well, that's because there isn't a tile for left and right walls that also don't have the shadow. Just a single example of many...

And finally, I started this project later than I should have, giving us only 3 weeks to work on it.

What To Do Next Time

Too many cooks in the kitchen! I'll have to work with just a single artist on my next project. I ended up doing about 1/4 of the graphics myself. I would like to get to the point where I do 0% of the graphics. Just like the artist would expect to do 0% of the coding. It just makes sense...

And I still need a musician. :)

1GAM April: Day 23 (Finale)

As of tonight, I am done with the 1GAM April project, Sprung Fever. Regretfully, it is identical to last night's build. I got caught up with other things tonight, and the artists working on the project were unable to complete the last remaining assets for the last day of the month.

However, this is only the end of the initial development cycle! And as new assets come in, and new ideas come up, they may very well find their way into Sprung Fever.

I'll have a full postmortem for the project in the coming days. In the meantime, enjoy the game! And I'll be starting a new project within the week. As well, I'll pick up where I left off with the new collision detection system for melonJS. That should be completed for 0.9.8 in about two months! And 0.9.7 is coming literally any day now... Things are really looking great so far!

That's it for now! See you in a few days!!

Monday, April 29, 2013

1GAM April: Day 22

Sprung Fever

(Clicky for Make Big)

Library is finished! And this is a very difficult one. I haven't been able to complete it yet. ;) But there's an ending screen if you manage to make it through!

So that gives us two stages. It's awfully short, but a lot of fun to play.

I've also added some new sound effects to the game, which adds to the overall experience. There sounds for are 8-bit foot steps, sweat drops, and a crash for picking up bad items. Speaking of the bad items, they also have a negative effect now, and will instantly drain half of your attention! D'oh!

The only things that's *really* missing for a complete game is the item images. Although having an intro and real ending would be nice, as well. :) With just one day left in April, maybe I will come up with something! See you tomorrow!

Sunday, April 28, 2013

1GAM April: Day 21

Nothing new to report on 1GAM today, either! Spent some time with melonJS documentation again, watched some videos, had a nice nap, and planning for my next projects... That's all!

Tomorrow I'll be back on April's 1GAM entry to finish up the Library. And if there's time, add an end sequence! Anything else left in the month will just go to bonus additions. :) See you tomorrow!

1GAM April: Day 20

Skipped a day to work on other stuff!

  • melonJS documentation overhaul
  • Planning and design on Ludum Dare entry (still unsure if I will enter)
Will be back on Sprung Fever tomorrow!

Saturday, April 27, 2013

1GAM April: Day 19

Sprung Fever

(Clicky for Make Big)

Library! Tonight I have started the Library! There are no girls, and the maze is only half complete. But it's a start. The three errands are all too easy, but that will be solved when the girls come in...

I'm also considering doing a new Ludum Dare project tomorrow! If I do, it will be a ridiculously small game (cramming a game into only about half of the available time... could be quite tricky!) But I think it will also help relieve some of the stagnation I'm starting to feel with Sprung Fever. (I love this game! But like all projects, this one is wearing on me.)

Perhaps also for my 1GAM May, I will limit myself to a much shorter development cycle. I have some other work that can be done, anyway. Mostly melonJS 0.9.8 stuff. :) We already have a pretty good collection of tickets lined up for it after the 0.9.7 release.

Final note; one of my artists is away on holiday for the last week of the month, so I will likely have to fill in the gaps for the final stretch. If I can come up with a third scenario that will be easy to draw, then I'll just do that. And then again, maybe the other artist can do some tiles and finish up whatever is left. That would be especially helpful!

Good night! See you tomorrow!

Thursday, April 25, 2013

1GAM April: Day 18

Sprung Fever 


Springy flowers! These are pretty cool actually :) The flowers are on "springs" drawn with a bunch of bezier curves that are animated with lots of trial-and-error. With a little variance on the cycle for animating the X/Y position, it gives a "springy" look. The flower rotates a bit as well which helps with the effect. No physics! Just a ton of "shortcut" maths for this one...

I also removed the yellow flowers from the background. They looked too distracting and ugly.

Finally, I also fixed the blurry errands note on iOS. I just upscaled the image by 1600%, because scaling down on iOS isn't nearly as ugly with the forced bilinear filtering.

Tomorrow I plan to work on the library (I need more tiles for this) and probably add some more girl variations. Mostly the new girls will just be palette swaps. :\ But I can give them each different AI, so different strategies will help while playing. It will be fun! See you tomorrow!

Wednesday, April 24, 2013

1GAM April: Day 17

Sprung Fever


Things are starting to slow down. :\ I guess it's just that time! It really is difficult to keep a fast pace for more than a few weeks.

Anyway, I added entity pooling as an optimization (may be noticeable on mobile?) And updated the title screen with some flowers... Not sure I like it as much as the plain grassy field.

I started on the flowers for the logo, too. But ran out of time for tonight. They will be animated and quite "springy" when complete! But until then... See you tomorrow!

Tuesday, April 23, 2013

1GAM April: Day 16

Play online

For today, I've fixed the mobile UI controls! It took a while to figure out exactly what was wrong with it, because it's not very intuitive. Basically, I have to keep track of every touch event by id that ever occurs, and iterate through that list, not the list provided by the event. That's because melonJS provides only the `changedTouches` list, and ignores the complete list of touches.

This sounds like a feature request here; include the `touches` list and `changedTouches` list like the spec! That would mean I won't have to track touches myself, just check `me.input.touches` in my `touchstart` handler, and `me.input.changedTouches` in my `touchend` handler. Done!

I also discovered in testing that mobile Safari is terribly slow on my iPhone because the game object count rapidly increases until the browser crashes. I haven't seen that on Chrome though, and I haven't verified it with CocoonJS yet. So that's going to need some investigation.

That's it for now! See you tomorrow!

Update: Fixed the object count problem. :) Actually, it's just a workaround within melonJS, but it's really helpful for this game (uses lots of Tweens!)

Monday, April 22, 2013

1GAM April: Day 15

Sprung Fever

(Clicky for Make Big)

Super quick update for tonight! I added a first iteration for mobile UI/touch controls. It has some bugs, but it gets the idea across! (The UI is only enabled for mobile devices.) I actually spent about an hour in Gimp making the buttons. They turned out acceptably.

I also discovered with this new feature that the game runs very poorly on iPhone 4S in Safari, but runs great on the same phone in CocoonJS! Way to go Ludei! Again. ;)

Final note: I also fixed a minor bug where the sweat particles would collide with the environment. Oops! Hehe!

Sunday, April 21, 2013

1GAM April: Day 14

Sprung Fever

(Clicky for Make Big)

Aw yeah! Day 14! Two weeks down! Just one more to go. (I really need to break that habit of starting late...)

Finally I've added a title screen, which you can see above. It could be way better with some artistic talent (or a lot of time) but I already spent hours on it (because I have no idea what I'm doing). So that's the end result! Hooray!!

I actually do want to make the title screen a bit more juicy. For starters, I want to add some flowers (being "spring" and all) and I'll probably stylize one or two and make the stems bouncy springs. That should be fun! But I've had enough title screen madness for one day...

Anyway, there are a lot of new goodies in this update that you'll find after the title screen! I continued working on things late into the night after yesterday's update to get all of this crammed in. So here's what's new:

  • Brand new HUD (semitransparent background, new meters based on Burcag's design, and a heart rate indicator)
  • Added the "redness" overlay nervous effect.
  • Added "sweat" particle effects as a nervous effect.
  • Tweaked all of the sound levels.
  • Added a "heartbeat" nervous effect.

The heartbeat nervous effect is probably my favorite thing in tonight's update! Play with headphones or a really good subwoofer. ;) The heartbeat sound effect is very bassy, so you won't hear anything with standard laptop speakers. But really, it's the entire ensemble of nervous effects that make it entertaining to avoid the girls while collecting items. I've fallen prey to panic on multiple occasions while testing this game. It's certainly the most fun to play of all my projects. (But more on that later!)

Ah, finally it's starting to look like a real game! This is the point of the Cousins Graph where the curve is starting to break away from "shit". And one of my favorite parts of the development cycle. :D

Saturday, April 20, 2013

1GAM April: Day 13

Play online

For Day 13 I've started on the second stage; The Library! I used abundant-music.com again to create the music, but didn't spend as much time on it. As a result, I think it's not nearly as catchy. (But hey! I'm not a musician!)

The other changes for today are; level change/restart transitions, and new door tiles by Matwek.

I also spent some time doing some behind-the-scenes work, like a very minor bug fix for melonJS (and pulling in other updates for it) and moving the errands list animation into the correct module. These changes don't affect gameplay at all, but it's worth mentioning how my time is spent on this project. (That's one of the goals of this blog, after all!)

I didn't get to the title screen work yet. I'll start on that shortly. And if I finish by tomorrow, it will even be the screenshot for Day 14! See you then!